Catherine Woodgold wrote:
Betty Woolf ) writes:
[excerpts from Betty's post]

Catherine Woodgold wrote:
snips
I don't think Straus et al have any more responsibility than
anyone else to prove something about spanking; they just happen
to be doing studies on that topic. I read recently that Straus
used to be a pro-spanker but changed his mind based on the results
of the various studies he's done over the years.

Well, people usually do studies trying to prove something, at least in
my experience. Spanking either has a positive effect, a negative
effect, or no effect on a child's behavior. My stance is colored by
what studies mean in my line of work, and you may not realize that the
way you state things has a particular meaning in the world of the harder
sciences. It is that that I am reacting to when I may seem to be
responding to things you think you didn't say; by my frame of
reference, I am responding to what I think you said.


The null hypothesis is that spanking
does no harm, not that it does good.


I'm not sure what particular statistical test
you're referring to here.

It's not a statistical test, but a basic principal within statistics.
You can't have statistical significance without a null hypothesis -
generally expressed as measuring the probability that whatever you've
observed could have occurred by chance. When I typed the above, I was
speaking of the null hypothesis as I use it in my job as a molecular
biologist. I went ahead and looked up the statistical definition.

Excerpts from http://en.wikipedia.org/wiki/Null_hypothesis

"In statistics, a null hypothesis is a hypothesis that is presumed true
until statistical evidence in the form of a hypothesis test indicates
otherwise.

If experimental observations contradict the prediction of the null
hypothesis, it means that either the null hypothesis is false, or we
have observed an event with very low probability. This gives us high
confidence in the falsehood of the null hypothesis, which can be
improved by increasing the number of trials. However, accepting the
alternative hypothesis only commits us to a difference in observed
parameters; it does not prove that the theory or principles that
predicted such a difference is true, since it is always possible that
the difference could be due to additional factors not recognized by the
theory."

Perhaps reading the above will help you understand where I am coming
from in analyzing these studies.


Spanking always causes pain, which is a form of harm.

One could substitute something else that causes pain into the above
sentence, such as "Falling off your bike always causes pain, which is a
form of harm." Yet no-one would suggest that kids should not learn to
ride a bike because of the pain involved in the inevitable falls.


I pretty much agree with the above. Do you agree or
disagree (or neither) with the statement "Spanking always
causes pain, which is a form of harm"?

I have to disagree, since I can remember spankings in which my reaction
was surprise rather than pain.


I think that the valid arguments against even mild spanking are that it
sends the wrong message and that it can be a betrayal of trust.


Those also strike me as valid arguments.


The "any pain is harm" argument does not convince me in the least.


I think it would be interesting to find out exactly where
this argument loses you. Do you agree that pain is harm?

Not always. One of my hobbies is horseback riding, and after riding I
am sometimes in pain, and I don't consider that bad.

I assume you're against painfully knocking kids off their
bikes for the fun of it. If so, why are you against it?
Because pain is harm, or for some other reason?

Because "for the fun of it" is not a valid reason. I can think of
reasons for physically stopping a bike in such a way that the rider
would fall: he is heading into traffic in a situation where he can't
hear me, or he is behaving maliciously and heading for a dog or small
child and I feel that is the only way to avoid more serious injury.

Why does
the same reason, whatever it is, not also apply to spanking?

Apples and oranges, to my way of thinking.


Betty Woolf said:

I obviously can't argue with people about what their intent is. It is
however possible to offer alternatives (which you do very well) without
playing the "studies say X causes harm." This is a hot button with me -
not directed at you - and the reason that I stopped reading
misc.kids.breastfeeding.


By the way, I also have other reasons for posting, including
that it's enjoyable and educational for me (and I hope for others),
and that I occasionally end up taking my own advice. :-)

Certainly it is possible to merely mention alternatives and
not discuss the reasons for doing one thing rather than another.
However, I don't choose to do so. I believe it's both interesting
and important to discuss the reasons for and effects of various
alternatives.

I'm not advocating not discussing the reasons for doing so, exactly. I
don't spank because I don't think it's effective, and I'm happy to share
what I think about alternative discipline methods if I think it will
help. I don't believe I know what's best for everybody in every
situation though.



Respondents can be very sensitive to the wording of the question and try
to guess the "correct" answer, even in anonymous surveys. I personally
have a terrible time with telephone surveys because I see shades of grey
everywhere and I could choose one of several answers. For example, if
the question is "Have you ever spanked your child?" the answer is no.
If the question is "Have you ever physically disciplined your child?"
the answer is technically yes - I have physically carried him places he
doesn't want to go, and physically restrained him from doing things he
wants to do.


I would think those sorts of effects would contribute to the
"noise" in the results. Usually this sort of "noise" tends to
cancel out when averages are taken over large numbers of respondents.
If a statistically significant correlation is found, there
has to be some reason for it (unless it's a fluke, which
becomes less likely as studies are replicated and depending
on the level of statistical significance). Until someone
comes up with some alternative explanation of the results,
it seems reasonable to suppose that they show a statistical
correlation between spanking and misbehaviour.

I don't think it's necessary to come up with an alternative explanation,
but only to point out things that weren't considered or flaws in the
methodology, as the wikipedia quote above indicates.


I'll look up the study when I can but I just don't see how you could ask
specific enough questions to reach a reasonable conclusion without
subconsciously leading the survey-taker to answer the way you want them to.


I think the survey questions were asked by professional
survey-takers who presumably didn't care what the answers
were. The wording of the questions was predetermined.

Right - it is the wording that I am wondering about.


Was the question whether the mother had ever spanked the child (as both
your original quote and the study title indicate) or whether the child
had ever been spanked at all, to the mother's knowledge?


"Mothers were asked how often in the past six months they had
'spanked, slapped or hit' the target child when the child 'does
something bad or something you don't like, or is disobedient.'"

And what about the argument that the one spanking, 8 months ago, was
*so* effective that there was no hitting in the last six months?


If it's the
former, I have to say that I have huge doubts about the validity of this
study. How can you draw valid conclusions from a study that disregards
the input/behavior of the father?


It's very simple. I think it's a very valid assumption that
children who are spanked by their mothers are, on average, spanked
more than children who are not spanked by their mothers.
In any case, if you don't believe that assumption, then the
study results can still be interpreted validly as showing
a correlation between spanking by mothers and misbehaviour.

On average, sure. But the methodology leaves a lot of unanswered
questions. What if the best behaved kids, never spanked by their
mothers, were products of the "Just wait until your father gets home"
model of discipline, or worse, that those kids not only were hit by
their fathers but watched as their fathers hit their mothers. Of course
I don't think that's true, but scientifically speaking, it's not ruled out.


I don't expect you to summarize the whole thing, but here are my
thoughts on the above, in case anyone is interested in the kinds of
questions these statements raise for me:

First, IMO "In the last 6 months" is a *huge* amount of time when you're
asking a parent to recall behavior, and second, the questions are so
vague that they can be misinterpreted easily.


Again, it seems to me that those factors would contribute to
the "noise" or experimental error and would tend to cancel out
when large numbers of results are averaged. Increased noise
generally reduces the chance that any correlations of any sort
will be found. If a correlation is found, it usually means
that there is some pattern which is strong enough to be
seen in spite of the noise. You haven't provided any explanation
of how the statistically significant correlations between
variables designed to measure spanking and misbehaviour
could have occurred if there is no actual correlation between
those variables. I'm not trying to imply that you have
any responsibility to provide such an explanation.
However, as long as no-one has come up with any plausible
alternative explanation, it's reasonable to interpret the
results as showing a correlation between spanking (by mothers)
and misbehaviour.

I promise I will try to look at the study, but it's not going to happen
soon. Off the top of my head, other factors I would want to know about
would be the following. I don't know if all or none were controlled for:

Income, marital situation, number of children, child's place in the
family (only, oldest, middle, youngest, step-children, and age
differences), sex of children, daycare situation, amount of sleep,
amount of exercise, alternative forms of discipline, and certain medical
information, such as diagnosis of ADD, autism spectrum disorders,
familial history of allergy, stuff like that.


Also, unless the mother
is with the kid 24 hours a day, she can't possibly answer with 100%
certainty.


She answered about her own spanking behaviour.

Right, but does she really know about the children's behavior when they
are away from her? Maybe the "low ASB" kids are bullying younger kids
with no fear of punishment, but in such a way that it is not called to
the mother's attention.


It is possible to estimate the amount of harm from a single
spanking by measuring the approximate average amount of harm
from a large number of spankings, and then dividing.

I don't believe that is valid.


Could you explain in more detail what you mean here?
Do you agree that it is possible to carry out such
a calculation?

No, I don't think it is possible to calculate. What is the unit of
measurement?

Can you suggest any better way to
estimate the amount of harm from a single spanking?

I don't think it's possible to estimate, so I have no answer here.

Parents need estimates of such harm to make ordinary,
day-to-day decisions, even if they don't think about
it in mathematical terms.

I don't agree with this either. I don't hit because I don't think it is
effective. I reached this conclusion long before I had kids, mostly
from working with animals. I never even thought about measuring the
harm of physical punishment, but rather observed over and over that it
doesn't work in most situations. The way we look at things with
training animals is "Set them up to succeed" and "make the right thing
easy and the wrong thing hard." Hitting has no place in that.

That is, I think that the fact that it is not effective, is, in-and-of
itself, sufficient to rule it out, without worrying about whether it
harms the recipient, because in most cases I don't think it has a
long-term harm.


I don't think so - I
think the argument is whether rare spanking causes measurable harm over
no spanking.


Correlations between spanking and misbehaviour have been
established in scientific studies. No-one has established
that any specific type of spanking (e.g. rare spanking)
is exempt from the same dynamics.

But is that because of study design or because of "truth"?

What do you believe? Do you believe rare spanking causes
harm? Do you believe it doesn't cause harm?

I believe hitting is not an effective way to positively influence
behavior. I don't know how to quantify "harm," and I don't know how to
separate the physical act of spanking from the controlling attitude of
parents who are likely to spank regularly. I lean toward thinking it is
not the actual spanking that does harm, but the adversarial relationship
itself, of which spanking is only one facet.

I see no reason to assume rare spankings would cause
no harm.

I think this is the crux of our disagreement. I see no reason to assume
rare spankings would cause harm, just as I see no reason to assume that
one cigarette or one Big Mac or one alcoholic drink would cause harm, on
average.

(oops, snipped too much)


What do you think? Do you think we just don't
know? Or do you think people can spank their
kids a small number of times with confidence that
each of those spankings will do far less harm than
each spanking would do to a child who was
accustomed to frequent spanking? Or what?

I think that we don't (and can't) know how much harm a spanking does in
the grand scheme of everything we do in raising our kids. I think that
even if we could measure the harm in a single spanking, that there are
things that can be done to negate that harm, in the context of the
larger relationship, just as there are health choices we can make to
minimize the damage from an ill-advised period of heavy smoking. And I
therefore object to even hinting that people who rarely hit their kids
are irreparably harming their kids and their relationship with their kids.


What is the evidence that spanking makes other discipline methods less
effective?

Well, think about it. What do you do when someone tries
to forcefully make you do something you don't want to do?
Most people react by digging in their heels. If that
same person later comes to you and tries to ask you
nicely to do that same thing -- or even something else --
how are you going to react?

Are you speaking purely of physical force?


No, it can be other forms of overt manipulation such
as punishment. The more forceful, severe and threatening
the punishment, the more a person tends to resent it
and to try to oppose further manipulation by the same person.

In some cases. But if I get a speeding ticket for going 5 miles per
hour over the speed limit and am threatened with license revocation if I
get another one, I'm going to probably slow down in addition to feeling
resentful.


I'm not going to comment much until I do read the studies but if you
start with "theoretical reasons to explain X" I guarantee that you can
come up with a study design that will support your conclusions.


Many people have tried to come up with a study design
which establishes a correlation between spanking and any
form of long-term benefit, but they have all failed.
This seems to be a counterexample to the idea that
you can find a study that shows whatever you want
(though I'm not sure whether that's what you meant).

I didn't say you could find one, I said you could design them. If I
wanted to refute the Straus et al studies I could design studies such
that his correlation would be lost in the noise. I could redefine
antisocial behavior. I think it's probably true that most of the
richest people in the US were probably spanked as children, just by
virtue of their age and background - I could most likely design a study
correlating rare spanking with earning potential.


I would also have to think twice before accepting ASB between the ages of
2 and 14 as "long term" misbehavior.


When I've said "long term" in this discussion I've
generally meant about 1 or 2 years or more. You can use
the phrase to mean whatever you want when you use it;
I hope you'll make the meaning clear enough in context.

I know this is going to sound very uncaring, but I'm not sure it matters
if spanking causes 2-to-14-year-olds to misbehave if they grow up to
be productive members of society.

Possible alternative explanations to "spanking causes misbehavior" -

-Children who are neurologically atypical are more likely to engage in
behaviors that are both antisocial *and* on the list of behaviors that
average parents will spank for.


This was a valid criticism of all the studies finding
correlations between spanking and misbehaviour until 1997.
In 1997, two important studies were published side-by-side
in a medical journal (Straus et al 1997 and Gunnoe and
Mariner 1997). Each of these studies looked at spanking
and misbehaviour varying over a 2- or 5-year time period.
The amount of increase or decrease in misbehaviour could
be measured. It was found that, on average, there was
more of an increase in misbehaviour in the kids who
were being spanked more at the beginning of the time
period, controlling for the amount of misbehaviour at
the beginning.

This is a correlation between spanking and misbehaviour
which could not be caused by pre-existing tendencies
of the child to misbehave.

This seems to refer to more frequently spanked subjects. I've already
said I accept a correlation between spanking as a primary discipline
method and misbehavior. I don't think you can use one study with
different methodology to support another study that didn't look at that
factor.


-The surveys are failing to even attempt to look for other explanations
and therefore don't question things that would provide alternative
answers.


That is simply not true. These studies carefully control
for a number of variables such as sex, socio-economic status,
a parental warmth variable and a number of other variables which
could otherwise confound the results.

I think there are always important variables that aren't considered. As
I said, I think it's too complicated.


-The study subjects are inappropriately chosen or classified.


How could that do other than contribute to the noise?
Actually, in Straus and Mouradian, the subjects were
chosen by dialing random telephone numbers. This seems
an appropriate method to find a random sample
of subjects.

Well, it guarantees that respondents had telephones, which lets out
people in some parts of the country and some socioeconomic classes. And
if they did the calls during the day that lets out families in which
both parents work. And if the call came at night there are assumptions
to be made about who has time to talk on the phone. Perhaps this also
classifies as noise.


I think we agree that spanking is not desirable or effective and that
parents should be encouraged to practice other discipline strategies.

I think we disagree on the possible harmful effects of a few spankings.


It would be interesting to find out exactly what the
disagreement is. I believe that spanking is harmful
whether it is done to a child who is used to it or not.
What do you believe?

I believe that the harm of a single spanking is not measurable, and that
even a few spankings, on average, have a negligible effect on the
child's long-term mental health, productivity, or ability to be a
contributing member of society.


I think we disagree about the place studies have in the discussion - I
don't trust them as you do.


Can you identify specific statements I've made that
you disagree with?

It's not a specific statement. I think it's two things: One is possibly
because I'm not as familiar with the literature, and that is that you
seem to be using conclusions from one study with a particular
methodology to support things from another study with a different
methodology. Depending on the details, that can be OK or misleading and
incorrect.

Second, with something like parenting, "studies say" and averages are
important to an extent but there are so many nuances that I hesitate to
use studies in the context of parenting discussions at all. For
example, there are sleep studies showing that "crying it out" is not
harmful, that young kids sleep better after CIO techniques are used,
etc. I absolutely don't care in that context about an average kid -
it's something that I could never ever do to my son.

Have I clarified, or muddied the waters?

That's all for tonight, in any case!


Straus M. A., D.B. Sugarman and J. Giles-Sims,
1997. Spanking by Parents and Subsequent
Antisocial Behavior of Children. Arch. Pediatr.
Adolesc. Med. Vol. 151, Aug 1997. pp. 761-767

Gunnoe, M.L., C.L. Mariner, 1997. Toward a
developmental-contextual model of the effects
of parental spanking on children's aggression.
Arch. Pediatr. Adolesc. Med. Vol. 151, Aug. 1997,
pp. 768-775.

Oh -- I just remembered: I think those last two studies are
available on the Internet! Oh, maybe only the abstracts, but
you can also find replies published in the same journal.
You can do a Google search on the name of the journal.
(Archives of Pediatrics and Adolescent Medicine.)
I just did and it works, but the URL is rather long.
I think they only supply the abstracts, though.
The full articles can be obtained via interlibrary loan
from public libraries.
--
Cathy Woodgold
http://www.ncf.ca/~an588/par_home.html
There are two types of people in the world: those
who divide the world into two types of people, and